I'm running 10.0.0 on a Win 8x64 host. I have a Win 7 guest. I mounted an image of a physical disk in my host. It's mounted as a SCSI disk and shows up as a choice in VMware. I selected SCSI as the disk type as well as non-persistent. VMware won't add the disk and reports that it "is already is in use. Failed to power on scsi0:1." I've added physical disks before, so has something changed with VMware 10? Thanks.
The disk geometrie looks a bit strange
Please replace your 6 geometrie lines with this 3
ddb.geometry.cylinders = "121601"
ddb.geometry.heads = "255"
ddb.geometry.sectors = "63"
I would also not declare the disk as rawDisk unless you really have to.
Do you really have to use Windows 8 ???
What type of Image?
How is it mounted?
Why not just manually make a Disk DescriptorFile for the Image without actually mounting the Image and add it to the VM and run it that way?
Thanks, Woody. I should have said that the image is an E01 (Expert Witness) format file. Otherwise, I would have created a vmdk descriptor. It is mounted as a physical disk and appears correctly in Windows Disk Manager. I could convert it to dd, but I rather not at this point.
Can the Image be mounted but no drive letter assigned and possibly offline and then add it to the VM? In other words as long Disk Management in the OS can see the disk, not necessarily present it for access to the User via File Explorer, I would think that WS would be able to see it and add it.
Doesn't work. I mounted the image as physical only and with a tool with which the disk doesn't appear at all in Disk Manager. The same failure/message appears.
Is the failure occurring during the Add Hardware Wizard? If yes then manually creating the Disk DescriptorFile and manually adding the .vmdk to the .vmx configuration file is what I'd try next. You know you can still manually create a Disk DescriptorFile even for the Raw Disk, so I'd assume one could also be created for the mounted E01 Image, assuming it's presented to the OS like, e.g: "\\.\PhysicalDrive1"
Thanks, Woody. I created the vmdk (doe-5.vmdk -attached) by just creating a new VM from the mounted image. Then I added these lines to the vmx:
scsi0:1.present = "TRUE"
scsi0:1.fileName = "doe-5.vmdk"
scsi0:1.deviceType = "rawDisk"
When I powered on the VM, I received the error message in the screenshot.
The disk geometrie looks a bit strange
Please replace your 6 geometrie lines with this 3
ddb.geometry.cylinders = "121601"
ddb.geometry.heads = "255"
ddb.geometry.sectors = "63"
I would also not declare the disk as rawDisk unless you really have to.
Do you really have to use Windows 8 ???
Thanks, Ulli. I had allowed VMware to calculate the geometry. If I had done it myself, I would use the figures you mentioned. I usualy remove the "raw" so that I can use snapshots. Anyway, I changed the vmdk to 121601-255-63 and changed rawDisk to Disk. Booting still failed, and I received the message in the attached screenshot. Note that the disk was mounted as Disk 10 this time, and I accounted for that in the vmdk. However, I then remounted the disk as writable, and everything worked. I have a tool that can mount a disk and cache writes, so that nothing actually is written to the disk. It basically just "fools" the operating system into thinking that writes were made.
JimmyW wrote:
I'm running 10.0.0 on a Win 8x64 host. I have a Win 7 guest. I mounted an image of a physical disk in my host. It's mounted as a SCSI disk and shows up as a choice in VMware. I selected SCSI as the disk type as well as non-persistent. VMware won't add the disk and reports that it "is already is in use. Failed to power on scsi0:1." I've added physical disks before, so has something changed with VMware 10? Thanks.
Changed with VMware 10 compared to which prior version?
Something was definitely changed/broken in VMware Workstation and Player at some point. I don't know exactly the Version where it broke in Workstation, but for the Player it was between 3.1.5 and 4.0.0. On VMware Players up to including 3.1.5, it is possible to mount physical disks as "raw disk", even if there were active partitions on that disk. Doing it is somewhat dangerous, because having two operating system mounting a partition writable will usually wreck that partition. But when trying to maintain or fixing boot-blocks of a multi-boot system (such as a multi-boot system where one of the Windows install uses TrueCrypt, whereas multi-boot linux uses Grub and other multi-boot Windows uses its default install (all on the first disk), being able to do this is extremely helpful -- especially when upgrading TrueCrypt (which can not reasonably cope with multi-boots). I don't mind if the dangerous mode of operation is not enabled by default, but with VMware Player 4.x this extremely useful feature seems to have been silently and completely removed.
Using "mount individual partitions" is *NO* suitable alternative, because this will divert access to the disk MBR (track 0) and partition boot blocks the -pt.vmdk file, rather than letting you access and modify the real boot blocks on the physical disk.
>Changed with VMware 10 compared to which prior version?
Compared with 9.x. I never had a problem with mounting disks with active partitions. As noted in my last post, I was able to add the mounted disk after making it writable.